PipeMIRSEQ: pipeline integrativo para análises de expressão diferencial em dados de MIRNA-SEQ de plantas

Abstract

Computational techniques have been highly used in the recent years to mitigate biological issues. Currently, many studies of non-coding RNAs (ncRNAs) using bioinformatics have been presented. The miRNAs are one of the ncRNAs classes and the use of bioinformatics is crucial for a rapid, cost-effective and reliable solution for analysis of these molecules, for instance, prediction of the precursors, mature, and target sequences; and expression analysis. For expression analysis, the use of tool s, platforms, and pipelines has been essential to analyze large -scale miRNA-seq data. However, there are few aiming at the analysis of plant miRNAseq. The aim of this work was to develop a pipeli ne (pipeMIRSEQ) to assist in an efficient and reliable way to all the data analysis steps comprised within miRNAseq. The implementation was separated in modules: quality analysis, using four kinds of tools; mapping of the reads, comparing three different aligners; quantification, considering the homolog miRNA families; differential expression, using two packages for calculation; and, at last, a review step, aiming at synthesizing all the information from each step. To validate the pipeline a reduced miRNA-seq dataset from coffee (Coffea arabica L.) was used. It was possible to observe that the pipeMIRSEQ could analyze all the dataset, presenting results comparable to the literature, and, therefore, achieving its aims. At last, improvements for the pipeline regarding both computational and biological aspects are discussed.